Useful and unuseful summaries of regression models Roberto

where logP is the studied experimental response and V1 and V2 are two independent variables for which a quantitative relationship with the response has been searched for. The numbers in the equation are called regression coefficients, being the first one the intercept of the regression model. n is the number of samples used for building the regression equation, r is the multiple correlation coefficient, sy is the residual standard deviation (or error standard deviation), F the calculated value of the F-ratio test. In several cases, r = 0.9765 is replaced by R = 0.9765 or by R = 0.9536 or by R = 95.36 %, being the last two indices (coefficient of determination) the squared r (R) value and the last one the corresponding percentage of the variance explained by the regression model. For evaluating the quality of the summary, mathematical definitions of the quantities involved in the previous example are given.